Image Retrieval Technique Based On MPEG-7

Along with the arrival of the multimedia age, more and more images were used widely. Content-based image retrieval technology has been a research hot spot in the recent years. On the other hand, due to the complexity of the information, the lack of standard for the description of image content induces low interoperability among CBIR systems. To meet this challenge, MPEG-7 was started as a solution to the normative description of audiovisual content. MPEG-7 relates the description with its content, and enables the effective and efficient query and access to multimedia content.In this paper, the main research objective is the study of image retrieval based on MPEG-7. Some significative conclusions are drawed from this study. The major contents include:The extraction of color descriptors is reported in detailed. The color layout and color structure descriptor are used in retrieval test and the performance is discussed.An extraction of dominant color is presented and it obtains better retrieval effectiveness compared with the extraction, which also baseon vector quantization. A method which adds the weight to blocks was applied to dominant color and the improved effect is obvious.A method for image retrieval based on structuring quantized histogram is presented. The color structure histogram and color histogram are the specific cases of structuring quantized histogram, which is endowed an extreme quantized step. The performances of the three type of histogram are discussed through the retrieval test and results show that the performance of structuring quantized histogram is the best.An actual CBIR system was developed based on dominant color and color layout, meeting global and local similarity needs. Retrieval results obtained from combined features are superior to that obtained from single feature.